Advantages and Disadvantages of Keeping a Defendant Incarcerated Prior to Trial

Advantages of keeping a defendant in jail until trial include the following: (1) the defendant is assured of being present for the trial and (2) the defendant will not be committing (possibly) additional crimes. (Figueroa (2018) notes that problems with illegal drugs are a major reason for defendants out of jail on bail to commit new crimes.) Disadvantages associated with keeping a defendant in jail until trial include: (1) a possibly innocent person will be kept in jail; (2) the defendant will not be able to continue working a job, and may lose that job as a result; (3) the defendant will have a more difficult time getting their affairs in order given that a guilty verdict and a jail sentence occurs; and (4) incarceration of a defendant is expensive—for example, the cost to New York City of incarcerating those who cannot afford bail prior to their trial is approximately $100 million annually (Stringer, 2018).

There are other disadvantages, at least to the defendant, of incarceration prior to trial. For example, defendants in pretrial detention are three times as probable to be sentenced to prison as those who can make bail (Lowenkamp, VanNostrand and Holsinger, 2013).

Given that tens of thousands of defendants can be (and typically are) affected by new laws which address pretrial conditions, it should not be surprising that bad outcomes can occur in each of the two cases: (1) reforms associated with more strict pretrial release conditions and (2) reforms associated with less strict pretrial release conditions.

For example, in July of 2015, Sandra Bland was stopped for a minor traffic violation in Waller County, Texas. A confrontation ensued and she was arrested and charged with assault. Unable to pay $500 in bail, Ms. Bland was found hanging in her jail cell three days after her arrest. Her death was ruled as a suicide, although there has been some dispute about this ruling (Hennessy-Fiske, 2015).

At the other end of the spectrum is the situation of Gerod Woodberry of New York City. Mr. Woodberry allegedly robbed banks in New York City four times: December 30, 2019; January 3, 2020; January 6, 2020; and January 8, 2020, before he was finally arrested on January 8. He was released the next day, on January 9, 2020, without bail under New York State’s new bail reform law. The main reason why he was released without bail is that his crimes were classified as robbery in the third degree and grand larceny but without the use of a gun; hence, his crimes were classified as nonviolent in nature. After being released on January 9, he allegedly robbed a fifth bank the next day, January 10, and was arrested again the same day. Mr. Woodberry was quoted as saying:

/ can’t believe they let me out. What were they thinking?

See Tracy (2020), Goldberg (2020), and Celona, Feuerherd, and Weissman (2020) for this and additional information on Mr. Woodberry’s case.

Current Approaches and Suggestions for the Bail Process

Often a judge will use a bail schedule as a guide in setting the bail amount. Some examples of typical bail amounts are shown in Table 3.1. These amounts are derived from (Bail Amounts by Crime..., 2019).

There is much variation in the amounts of money required for bail for a specific crime, depending on the jurisdiction. For example, for lower-income states like Utah, the bail amount for illegal possession of a controlled substance could be set as low as $500.

It has been reported that 40% of adult Americans do not have enough savings to cover a $400 emergency (Backman, 2018). For

TABLE 3.1

Typical Amounts for Bail for Various Crimes

Misdemeanor

arrested Americans, one could expect this percentage to be much higher. One can therefore see why paying the money required for bail for a crime would be so difficult for many people.

In deciding about pretrial conditions for a defendant such as whether to keep said defendant incarcerated or not incarcerated, the amount required for bail and other restrictive conditions, a judge may consider any of several characteristics of the situation. Some of these characteristics have already been mentioned above in the discussion of the Criminal Justice Reform Act of New Jersey and the Money Bail Reform Act of California. These characteristics include the alleged crime of the defendant, community safety, the likelihood that the defendant will appear for trial, the likelihood that the defendant will obstruct justice, the likelihood that the defendant will commit another crime while out on bail, the likelihood of the defendant committing a technical violation related to his/her pretrial conditions, history of drug abuse, prior criminal history, employment and housing situations of the defendant, zip code of defendant’s residence, nature of active community supervision at the time of arrest, the defendant’s ability to pay, and the amount of congestion in the local jail (see Van Nostrand & Keebler, 2009; Bail Amounts by Crime..., 2019; Simon. 2018; and Rahman, 2019).

Obviously, some of these characteristics are more important than others in the determination of pretrial conditions for the defendant.

In addition, some of these conditions are dependent upon each other in a probabilistic sense; for example, prior criminal history and the likelihood that the defendant will commit another crime while out on bail are certainly dependent.

In most jurisdictions, judges will consider the situational characteristics listed above in a subjective manner through what might be termed as a subjective risk assessment. In a few jurisdictions, more formal approaches are used such as described above in the Money Bail Reform Act of California, or as suggested with objective risk assessments (Cooprider, 2009). Whether a less formal or a more formal approach is used, certain principles should be followed in determining pretrial release conditions, as noted by Hopkins, Bains, and Doyle (2018):

• 1. Releasing defendants on their own recognizance should be the norm;
• 2. Burdens and restrictions on defendants should not exceed the true interests of the government;
• 3. Restrictions placed on the defendant should be based on the alleged crime, the likelihood of flight, and the pre-arrest criminal activity of the defendant;
• 4. The defendant should not be charged for pretrial services; and
• 5. Constraints on the defendant should be evidence-based.

A summary of the situational characteristics considered by various acts/laws is shown in Table 3.2.

TABLE 3.2

Summary of Situational Characteristics Considered by Judges in

Setting Pretrial Conditions

Source

2014 Criminal Justice Reform Act of New Jersey

Money Bail Reform Act of California

Bail Reform Law of New York State

Characteristics

community safety, probability that defendant will appear fortrial given that he/she is released, probability that defendant will obstruct justice if released

alleged crime of defendant, prior criminal history, employment and housing situations, zip code of residence, community safety

defendant's ability to pay

See Summers and Willis (2010) for additional research on pretrial risk assessment.

A Multiple Objective, Decision Theoretic Model for Selection of Pretrial Conditions

In this section, we present a model and an illustrative example of that model for the selection of pretrial conditions for a defendant. The model is based on the optimization of expected utility, where the utility function contains three attributes; one of the attributes represents the interests of the “government” (or “the people”) while the other two attributes represent the interests of the defendant. The fact that two of the attributes are mainly of concern to the defendant and only one is mainly of concern to the government is not an indication that the government’s concerns are of less importance than the defendant’s concerns. The weights associated with the utility function affect the overall value structure for the decision and can result in the government’s concerns being of more importance than the defendant’s concerns, even with one attribute versus two.

Uncertainty, and its inherent risk, associated with the behavior of the defendant is represented through a Monte Carlo simulation of the situation.

The model is to be used for a single defendant. However, it could be easily extended to consider a whole group of defendants, as would be done when considering a new law for the situation.

Stakeholders

Decisions regarding pretrial conditions for defendants should be based on probabilities, benefits, and costs that are associated with specific outcomes. In addition, the various trade-offs between multiple performance measures should be considered.

For example, in deciding whether or not to release a defendant prior to trial, and, if released, the amount of bail required and which restrictive conditions, if any, to place on the defendant, the judge will at least implicitly consider things like the (1) respective probabilities associated with the defendant committing additional crimes, intimidating witnesses, and not showing for trial; (2) the financial situation of the defendant; (3) the defendant’s prior criminal history; and (4) the defendant’s history of drug abuse among other things. With respect to the set of probabilities noted in (1), Bechtel, Holsinger, and Lowenkamp (2017) note that none of the conditions of release from their review considered the probability associated with the event of re-arrest of the defendant.

In the model and illustrative example developed in this section, we consider two categories of stakeholders: (1) the government/state and (2) the defendant. The category of the government/state includes any direct victim(s) of the crime and the taxpayers of the jurisdiction; typically, in criminal proceedings, this category is referred to as “the people”. The defendant refers to the person accused of the crime as well as the group of people directly connected to the defendant, such as immediate family. One thing to keep in mind regarding the defendant is that he/she may very well be innocent of the crime; hence his/ her interests (including the interests of direct family) should be considered in the decision regarding pretrial disposition.

Each of the two stakeholders are concerned with one or more performance measures (or attributes). We use a multiattribute utility function to represent the trade-offs that should be considered among the performance measures by a “super stakeholder” (i.e., the judge who considers both the government/state and the defendant). Remember that such a multiattribute utility function considers both the multiple, conflicting, performance measures as well as the uncertainty and resulting risk of the situation.